INSERT

Explanation

INSERT adds new rows to a table. You specify the target columns and the values for each new row. If you omit a column, it gets its default (or NULL if allowed).

There are three common shapes:

• One row: INSERT INTO t (cols) VALUES (...)
• Many rows: multiple VALUES tuples in one statement
• From a query: INSERT INTO t (cols) SELECT ... to copy filtered data

Syntax

INSERT INTO products (name, category, price, stock)
VALUES ('Desk Mat', 'Stationery', 11.99, 200);
 
-- Multiple rows:
INSERT INTO products (name, category, price, stock)
VALUES
  ('Webcam Cover', 'Electronics', 4.99, 500),
  ('Pen Refill', 'Stationery', 2.49, 800);
 
-- From a query:
INSERT INTO products_archive (name, price)
SELECT name, price FROM products WHERE stock = 0;

Interactive Example

Because this sandbox is read-only you can't persist inserts, but you can preview the exact row an INSERT would create using SELECT with literal values.

Common Mistakes

• Mismatched column/value counts. The number of columns must equal the number of values, or the statement fails.
• Forgetting NOT NULL columns. Omitting a column that disallows NULL and has no default triggers an error.
• Relying on column order. Inserting without naming columns depends on the table's physical order — brittle and error-prone.

Best Practices

• Name every column you're inserting to keep queries robust.
• Batch multiple rows into one INSERT for big loads.
• Use INSERT ... SELECT to copy data between tables in a single step.
• Wrap bulk inserts in a transaction so you can roll back on error.

Practice Question

Write an INSERT that adds two new products rows (a 'Furniture' item priced 99.99 with stock 50, and a 'Stationery' item priced 3.99 with stock 300), naming all four columns explicitly.

Summary

INSERT INTO adds rows, with columns named explicitly for safety. Insert one or many rows, or copy from a SELECT. Always match column count to value count and respect NOT NULL constraints.

FAQ